They swapped out MIT [0] for BSD-3 just five months ago, which doesn’t seem to have been announced at all. Nixpkgs still has the 0.6.5 release as MIT, and 0.6.6 is in the AUR as MIT as well.
I wanted to be on the dev’s side here. CLA’s can be reasonable and I’ve happily let others convert my GPL code to them before. But reading the Open WebUI developer’s blog [1] makes it evident this isn’t really about the community — as he says, “It’s just me” and what he wants that matters for Open WebUI.
Their Cloud ToS forbids sharing performance information (section 3.3i), which in its most generous interpretation is meant to avoid the type of speculation in this thread. https://www.atlassian.com/legal/cloud-terms-of-service
> But of the 850 people in the jail system that the EFF is talking about here, that means at least 4 people are innocent
Saying "at least 4" is pretty misleading here, since you're assuming the 0.5% is perfectly distributed. That estimate could be accurate without a single innocent in these specific 850.
It's not that misleading even if the distribution is not perfect. It would be extremely unlikely that not a single person is innocent.
Given that this is a jail and not a prison, these people are mostly awaiting trial. From a certain point of view (innocent until proven guilty), most of these people are innocent and they can't even get a letter.
True, but I think that misses the point. I think the comment was an attempt at facilitating empathy for the incarcerated who are in this situation since for most people they don't see themselves as criminals (even those who are in fact criminals) and it is harder to empathize with those who you see as being part of a different tribe/group/class.
The fact is that these are fellow humans who are suffering, not because we decided as a society that this was what justice means, nor because this was the result of some study that figured out this would decrease recidivism, but just because some for profit institution decided they wanted to do something that would directly result in them suffering. That is enough for me to support this lawsuit.
The "5.8% more likely" statistic would be in comparison to the population rate, not the subgroup who aren't overweight or obese. That group would have a 1-(1-.78)/(1-.736) or 16% reduction in hospitalization rate in comparison to the population, resulting in a relative increase of ~27% if you compare these two groups.
I would describe that heuristic as Bayesian, since the subject evaluates the prior evidence for each option, and then chooses the one with the most supporting information, instead of considering the options' logical or mathematical properties.
NNs are not binary -- they're fundamentally analog, so for classification problems, they produce probabilities that a test case is in each possible class. A binary "X/not-X" test often comes from applying a threshold to the NN's output.
Quoting from the Universal approximation theory's Wikipedia, "neural networks can represent a wide variety of interesting functions." While they may be much better at pattern recognition, it's possible to produce almost anything with one, including Go moves, if you can devise a method to interpret the outputs.
NNs are not binary -- they're fundamentally analog, so for classification problems, they produce probabilities that a test case is in each possible class.
I think that's "true and false". Yes, NNs produce a "chance of being X" but no, NNs don't naturally produce a human-intuitive "degree of being X". If the NN is looking for a "red fire engine", it's not necessarily going to produce "degree of redness" as its output.
-- As to neural nets representing many functions; neural nets, Taylor series, Fourier series and so-forth can represent/approximate "any function". True but effectively irrelevant. What matters is what function a given methodology can be trained, programmed or whatever into representing.
> from the perspective of security, usability, and especially structured data transfer it's terrible.
Email is basically free, works well for up to 5 megabytes of data, and data security isn't much of an issue for open source work. The post suggests quite a few tools that improve the Git-email workflow, and I think some do prefer those to certain web-hosted Git interfaces.
> there's plenty of standard ways to transmit data that aren't SMTP
Are they free, federated, and as reliable as email? It may be inferior in some technical ways, but it's still a rational choice for small non-private data transfers, such as a Git patch or any another text.
> data security isn't much of an issue for open source work
Is that the case? It seems like you may be focusing on specifically the privacy aspect of "security". I'd say that email is equally bad at ensuring integrity and authenticity, which are crucial aspects of security for open source work that's consumed by others. We can attempt to backfill those gaps in email using GPG and other tools, but we're trying to put a bandaid over a mortal wound in a lot of ways. Recent vulns have highlighted what has been known for a while: trying to ensure the authenticity and integrity of a protocol as broad as email with as much client-side complexity is a losing battle.
But then, that applies ten-fold for anything that uses HTTP, or god forbid, browsers. Just look at how even the matrix spec manages to be incompatible with the HTTP spec.
That post isn't explicitly about a Firefox Monitor extension, and I could only find this rollout to US users mentioned on 3rd party websites. Those posts include BugZilla links, but they're restricted.
I can easily disable Pocket, Autofill, and Screenshots in about:config, and Firefox support explains that. With this extension, I'm missing a "extensions.fxmonitor.enabled" option.
I understand it's good to build in optional security features like this, but I don't see how it's acceptable to not notify users or provide an opt-out.
They are missing the about:config setting.. but it's early. Might be a bug, or maybe its not being used(?)
Eventually it'll display a popup in the UI that will notify you if the website you're on has been breached. I think we can all agree that would be useful.
So reading that ghacks article, it goes live to all EN-US users on the 25th Sept (my browser already has it), but we can't switch it off yet via: about:config?filter=extensions.fxmonitor.enabled
Pretty poor to roll it out without the option of disabling it, especially as I also ticked no tracking.
Sure, it's a useful feature, but my password manager (1Password) already does it. I'm OK with testing out this extension regardless, but wouldn't be if I wanted to limit my data use like the post's author.
I wasn't a fan of the UI refresh for the same reason, so I made my own alternate homepage which sorts the feed by artist and de-emphasizes reposts (http://gurlitz.org/soundcloud-digest/). Unfortunately like you said, content also dried up around the same time.
I wanted to be on the dev’s side here. CLA’s can be reasonable and I’ve happily let others convert my GPL code to them before. But reading the Open WebUI developer’s blog [1] makes it evident this isn’t really about the community — as he says, “It’s just me” and what he wants that matters for Open WebUI.
[0]: https://github.com/open-webui/open-webui/pull/8468/files [1]: https://jryng.com/thoughts/my-purpose
reply